Agentic AI systems—autonomous, decision-making AI agents—are transforming cybersecurity by enabling both advanced defensive capabilities and increasingly automated cyberattacks. In critical infrastructures such as power systems and smart grids, these agents can enhance resilience through rapid detection and response, while also being misused for large-scale reconnaissance, coordinated attacks, and cyber-physical disruption. These dual-use capabilities raise urgent governance challenges, including questions of accountability, human oversight, escalation control, and cross-border responsibility. Using energy infrastructure as a case study, this talk examines how agentic AI reshapes the cybersecurity threat landscape and argues for international governance frameworks that can balance autonomy, security, innovation, and societal trust.
About the Speaker
Abbas Yazdinejad, Ph.D., is an Assistant Professor in the Department of Computer Science at the University of Regina and the Director of the Decentralized Cybersecurity & Artificial Intelligence Lab (DCAILab). He has been recognized among the World’s Top 2% Scientists (Stanford University ranking). His research focuses on artificial intelligence and machine learning for critical, safety- and security-sensitive applications, with particular emphasis on agentic AI systems, large language model (LLM) security and governance, autonomous cybersecurity, privacy-preserving machine learning, federated learning, Internet of Things (IoT) and Industrial IoT, and quantum computing. He has held postdoctoral research positions at the University of Toronto and the University of Guelph. His work bridges technical innovation, teaching, and industry-engaged research to advance trustworthy and resilient digital systems.
